ABSTRACT
Brain-machine interfaces to classify basic responses, "Yes" or "No," are particularly beneficial for individuals with amyotrophic lateral sclerosis (ALS) who have experienced loss of motor function. However, it is challenging to elicit these cognitive thoughts with high performance because of significant individual differences in brain activity. To overcome this challenge, a "sensory association paradigm," which connects sensory stimulation and Yes/No recall, has been introduced. In this study, we explored the potential of employing a genetic algorithm (GA) to reduce analysis time, with the goal of applying this paradigm at the bedside of ALS patients. We utilized galvanic vestibular stimulation (GVS), which causes equilibrium distortion, and coupled it with Yes/No responses through classical conditioning. Electroencephalographic (EEG) signals were recorded when the distortion occurred due to Yes/No recall alone in the absence of GVS, and machine learning was used to classify Yes/No responses. The classification accuracy was compared between the signals of EEG electrodes selected based on the brain activity areas by the distortion of equilibrium (area-selected) and those of electrodes selected by GA (GA-selected), and the GA-selected condition showed significantly higher accuracy than the other conditions. The selected electrodes were relevant to GVS elicitation and cognitive thinking, and the time required for analysis was within 0.1 seconds, indicating that the time required for bedside analysis may be sufficient. Additional features beyond the basic phase locking value (PLVs) used in this study will be incorporated in our future study to enhance the accuracy.
- Natsue Yoshimura, Kaito Umetsu, Alessandro Tonin, Yasuhisa Maruyama, Kyosuke Harada, Aygul Rana, Gowrishankar Ganesh, Ujwal Chaudhary, Yasuharu Koike, and Niels Birbaumer. 2021. Binary Semantic Classification Using Cortical Activation with Pavlovian-Conditioned Vestibular Responses in Healthy and Locked-In Individuals. Cerebral Cortex Communications 2. 7 (2021). 1-13.Google Scholar
- Chea-Yau Kee, S.G. Ponnambalam, and Chu-Kiong Loo. 2015. Multi-objective genetic algorithm as channel selection method for P300 and motor imagery data set. Neurocomputing 161. 3 (2015). 120-131.Google Scholar
- Luo Y, and Lu BL. 2018. EEG Data Augmentation for Emotion Recognition Using a Conditional Wasserstein GAN. 2018 40TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC). 8 (2018). 2535-2538.Google Scholar
- Ivan P. Pavlov. 1927. Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex. Oxford University Press. (1927).Google Scholar
- Richard C Fitzpatrick, Jon Marsden, Stephen R Lord, and Brian L Day. 2002. Galvanic vestibular stimulation evokes sensations of body rotation. Neuroreport vol.13 (18). 12 (2002). 2379-2383.Google Scholar
- Shiyu Luo, Miguel Angrick, Christopher Coogan, Daniel N. Candrea, Kimberley Wyse-Sookoo, Samyak Shah, Qinwan Rabbani, Griffin W. Milsap, Alexander R. Weiss, William S. Anderson, Donna C. Tippett, Nicholas J. Maragakis, Lora L. Clawson, Mariska J. Vansteensel, Brock A. Wester, Francesco V. Tenore, Hynek Hermansky, Matthew S. Fifer, Nick F. Ramsey, Nathan E. Crone. 2023. Stable Decoding from a Speech BCI Enables Control for an Individual with ALS without Recalibration for 3 Months. Advanced Science. 10 (2023).Google Scholar
- Jian WJ, Chen MY, and McFrland DJ. 2017. Use of phase-locking value in sensorimotor rhythm-based brain-computer interface: zero-phase coupling and effects of spatial filters. Medical & Biological Engineering & Computing 55(11). 11 (2017). 1915-1926.Google Scholar
- ZH, Zhang HM, Sun LN. 2023. An Auxiliary Synthesis Framework for Enhancing EEG-Based Classification With Limited Data. IEEE Trans Neural Syst Rehabilitation Engineering (31). 6 (2023). 2120-2131.Google Scholar
- Ikegami S, Takano K, Kondo K, Saeki N, and Kansaku K. 2014. A region-based two-step P300-based brain-computer interface for patients with amyotrophic lateral sclerosis. Clinical Neurophysiology. 11(2014). 2305-2312.Google ScholarCross Ref
Recommendations
Brain-Computer Interface using Directional Auditory Perception
AHs '23: Proceedings of the Augmented Humans International Conference 2023We investigate the potential of brain-computer interface (BCI) using electroencephalogram (EEG) induced by listening (or recalling) auditory stimuli of different directions. In the initial attempt, we apply a time series classification model based on ...
Brain--computer interface (BCI): is it strictly necessary to use random sequences in visual spellers?
APCHI '12: Proceedings of the 10th asia pacific conference on Computer human interactionThe P300 speller is a standard paradigm for brain--computer interfacing (BCI) based on electroencephalography (EEG). It exploits the fact that the user's selective attention to a target stimulus among a random sequence of stimuli enhances the magnitude ...
Adaptive classification for brain-machine interface with reinforcement learning
ICONIP'11: Proceedings of the 18th international conference on Neural Information Processing - Volume Part IBrain machine interface (BMI) is an interface that uses brain activity to interact with computer-based devices. We introduce a BMI system using electroencephalography (EEG) and the reinforcement learning method, in which event-related potential (ERP) ...
Comments